Abstract
Background
Spondyloarthritis (SpA) has specific features in women and may be triggered by parturition. The peripartum period is marked by musculoskeletal remodelling and IL-17 secreting cell activation to enable childbirth. We therefore hypothesized that peripartum SpA might predominantly involve the entheses and be less dependent on HLA-B27 and TNFα.
Methods
We conducted a retrospective, observational, case-control study using the Rouen University Hospital’s clinical data warehouse. Delivery dates and symptom onset were available for 154 women with SpA. Characteristics were compared between cases (with peripartum SpA) and controls.
Results
We identified 58 cases and 96 controls. HLA-B27 (37.3% vs. 52.3%, p-value: 0.04), anterior chest syndrome (74.1% vs. 54.7%, p-value: 0.012 and psoriasis (37.9% vs. 16.8%, p-value: 0.004 were significantly associated with peripartum SpA. Lower rates of biological inflammatory syndrome at TNFα inhibitor introduction, (22.2% vs. 48.1%, p-value: 0.003), of radiographic sacroiliac damage (4% vs. 18.6%, p-value: 0.012 and of MRI sacroiliitis (30.6% vs. 47.4%, p-value: 0.046 were observed in women with peripartum SpA. Also, a lower retention rate of TNFα inhibitors was observed in women with peripartum SpA, 45.6% at 12 months, versus 71.2% controls (Odd ratio: 2.56 [95% CI: 1.34–4.95]). Multivariate survival analysis found a hazard ratio for TNFα inhibitor discontinuation of 2.43 (95% CI: 1.62–3.65) within the first two years of treatment.
Conclusions
Our study highlights a distinct phenotype of peripartum SpA, and a lower retention rate of TNFα inhibitors suggesting a worse response to these class of bDMARD.
Trial registration
As we conducted a retrospective study without health care intervention, no prospective registration was done.
Keywords: Spondyloarthritis, TNFα inhibitor, Parturition, Peripartum
Introduction
Spondyloarthritis (SpA) is a rheumatic disease characterised by inflammatory involvement of the entheses [1]. The prevalence of SpA in Europe is 0.5% [2], with a balanced sex ratio for axial forms without radiographic involvement but sex differences (male to female ratio 2:1) in peripheral or axial forms with radiographic involvement [3].
The inflammatory process of SpA begins in the entheses, and is dependent on mechanical stress and immune activation [4, 5]. The IL-17-secreting cells most commonly found in peripheral joints are mast cells [6, 7], neutrophils [7] and Tγδ lymphocytes [8]. Macrophages and CD8 + T lymphocytes, TNFα-producing cells [9], have been identified in numerous human axial samples [10] as well as in peripheral samples [11].
In women, compared to men, different authors observed a more severe involvement of the entheses [12, 13], a weaker association with HLA-B27, a less elevated CRP level [14, 15], and a stronger association with inflammatory bowel disease (IBD) and psoriasis, but a weaker association with uveitis. Imaging modalities found less structural damage on X-rays [14, 15] and more sacroiliac joints oedema on MRI in patients without proper SpA [16]. Women also had a poorer response to anti-TNF drugs [13, 17].
In women, 18–28% of onset of SpA occurs in the postpartum period [18–20] but no recent studies have updated these rates. Disease activity increases in the postpartum period in 30–50% of SpA [21, 22] and appears to return to normal one year after delivery [23]. From the 32nd week of amenorrhoea [24], to enable childbirth, modifications will progressively occur in the female pelvis driven by sexual and non-sexual hormones such as relaxin [25, 26] associated with the secretion of metalloproteinases targeting type I (main constituent of tendons and ligaments [27]) and III collagens [28]. These modifications, responsible for microlesions in these tissues, also involve the entire musculoskeletal system. Parturition involves the activation of macrophages, neutrophils, mast cells [28] and Tγδ lymphocytes [29], which produce pro-inflammatory mediators such as IL17 [30]. TNFα is minimally expressed by decidual macrophages [31]. We therefore hypothesized that SpA occurring in the peripartum period might predominantly involve the entheses and be less dependent on HLA-B27 (as immune activation is already occurring) and TNFα.
To our knowledge, there is no cohort investigating the characteristics of women with SpA occurring in the peripartum or postpartum period. The aim of this case-control study was to investigate the characteristics of SpA occurring in women in the peripartum period and their response to treatment.
Materials and methods
Design
This single center, retrospective, observational and case-control study was conducted at Rouen University Hospital, France.
Setting
Patients were identified using EDSaN, Rouen University Hospital’s clinical data warehouse, which aggregates different fragmented healthcare data in French language from clinical information systems. More precisely, the system is able to search into unstructured data (e.g. clinical notes) dealing with nuances of natural language (e.g. negations) to maximize precision and recall [32]. We searched the queries (in French) « spondyloarthritis - ankylosing spondylitis - spondylarthritis - spondylarthropathy - sacroiliitis - psoriatic arthritis - SpA » with « pregnancy - postpartum », in all documents associated with consultations and hospitalizations from January 2000 to January 2023, containing rheumatology and obstetrics data. Files where symptom onset period and parturition date were mentioned by physicians, and that met our inclusion and exclusion criteria were then manually selected (n = 154).
Study inclusion criteria
Inclusion criteria were women over 18 years of age at the time of symptom onset, diagnosed with SpA by a rheumatologist experienced in inflammatory diseases at a tertiary care centre between January 2000 and January 2023. The diagnosis of SpA had to be validated at the last follow-up visit. Exclusion criteria were the association with another inflammatory rheumatism, women meeting the 2016 ACR fibromyalgia diagnostic criteria [33] at diagnosis, onset of symptoms after voluntary termination of pregnancy or during pregnancy before the 36th week of amenorrhoea, or unavailability of period of symptom onset and delivery. We excluded 282 patients for missing data about SpA onset or parturition, 113 for final diagnosis other that SpA or fibromyalgia criteria at diagnosis, 24 for diagnosis of SpA before the year 2000, 13 for other inflammatory disease associated with SpA and seven for onset of disease during pregnancy or after abortion.
Patients’ characteristics
Socio-demographic, clinical and biological characteristics were collected. Demographic characteristics included age at onset of symptoms, and smoking status. Clinical characteristics were collected from standardized observations and medical files and included family history of SpA, clinical involvement of sacroiliac joints, of spine, history of arthritis, enthesitis (anterior chest wall pain, calcaneal enthesitis, enthesitis at any site; entheses were clinically checked), dactylitis, arthralgia, uveitis, IBD, psoriasis, non-steroidal anti-inflammatory drug (NSAID) intake. Biological characteristics included history of biological inflammation (defined as a C reactive protein (CRP) level of > 5 mg/L), CRP at diagnosis, CRP at introduction of TNFα inhibitors, and HLA-B27 status. In addition, imaging data were collected from clinical reports: structural damage on X-ray (which were carried out at the clinician’s discretion during follow-up), and sacroiliitis on MRI (evaluation at diagnosis). Sacroiliitis on MRI was defined following the ASAS recommendation [34], and equivocal cases were jointly reviewed by a rheumatologist and a radiologist. Treatment response was assessed according to the following parameters: retention rates of the TNFα inhibitors, etanercept, infliximab, adalimumab, golimumab, and certolizumab, and Bath Ankylosing Spondylitis (BASDAI) and Bath Ankylosing Spondylitis (BASFI) scores before and 6 months after the introduction of TNFα inhibitors. Treatments of less than one month were not analyzed, and data were censored after 24 months of follow-up. We determined the proportion of women meeting the following classification criteria: axial and peripheral ASAS classification criteria [35], New York criteria [36], Amor criteria [37] and ESSG criteria [38].
Cases and controls
Cases were defined as women with onset of symptom in the period one month before delivery (in agreement with the onset of immunological and collagen remodelling mechanisms) and one year after delivery (in agreement with the date of return to basal activity of postpartum SpA flares) (n = 58). Women whose symptoms occurred outside this period were included as controls (n = 96), so they were either nulliparous women or develop symptoms 1 year after delivery. Treatment dosages followed the recommendations of the French regulatory authorities.
Analysis
Statistical analyses and graphs were generated using Python (Python Software Foundation - packages: scikit learn 1.1, matplotlib 3.6.0, seaborn 0.12.1, lifelines 0.27.7, sksurv 0.21.0). Quantitative variables were compared using Mann-Whitney or t-test (depending on distribution and number of patients to analyze) and qualitative variables using Boschloo tests. Survival analyses at selected time points were performed using log-rank test, after a log(-log(x)) transformation to increase statistical power [39]. We performed the multivariate survival model using Cox regression on variables pre-selected from the literature (CRP at TNFα inhibitor introduction and sacroiliitis on MRI [40]). P-values below 0.05 were considered significant, except for the multivariate analysis where Bonferroni correction was applied and a threshold of 0.017 was used.
Ethics approval
Our study was conducted in accordance with the Declaration of Helsinki and approved by the local ethics committee (CSE – EDSaN) for research on existing data at Rouen University Hospital (2021/0190/OB), approval need was waived by this comity. Moreover, the EDSaN data warehouse has been approved by the CNIL French agency (DT-2020-007, 19th october 2020).
Results
Comparison of clinical characteristics between women with spondyloarthritis in the peripartum period and controls
In total, 154 women were diagnosed with SpA in our centre between January 2000 and January 2023. Among them, 58 women with SpA occurring in the peripartum period were included as cases and the remaining 96 women were included as controls. Their demographic characteristics are summarised in Table 1. The mean age at disease onset was slightly but significantly higher in the peripartum group compared with the control group (28.7 ± 5.1 years versus 26.1 ± 6.8 years respectively). The delay to diagnosis was longer in the peripartum group compared to controls, 40.9 ± 50.4 months versus 28.5 ± 31.1 months, respectively. A lower rate of HLA-B27 was observed in the peripartum group compared to controls, 37.3% versus 52.3%, respectively (p-value: 0.04), with no difference in family history of SpA. A majority of women had axial pain, spinal pain, and peripheral arthralgia. There was a trend for more enthesitis in the peripartum group compared to controls, with a significant difference in anterior chest wall pain, 74.1% versus 54.7% respectively (p-value: 0.012). A higher rate of psoriasis was observed in the peripartum group compared to controls, 37.9% versus 16.8%, respectively (p-value: 0.004).
Table 1.
Demographic, clinical and biological characteristics of the 154 women included
| Women with SpA in the Peripartum period (n = 58) | Controls (n = 96) |
All patients (n = 154) | Missing data | p-value | |
|---|---|---|---|---|---|
| Age, mean ± SD (years) | 28.7 ± 5.1 | 26.1 ± 6.8 | 27.1 ± 6.3 | 3 (1.9%) | 0.013 |
| Smoker, n (%) | 15 (44.1) | 40 (58.8) | 55 (53.9) | 52 (33.8) | 0.174 |
| HLA-B27, n (%) | 19 (37.3) | 47 (52.3) | 66 (42.9) | 14 (9.1) | |
| Family history of SpA, n (%) | 14 (29.2) | 23 (29.1) | 37 (29.1) | 27 (17.5) | 1.000 |
| Nulliparous women at symptom onset | 0 (0.0) | 74 (77.1) | 74 (48.1) | 0 (0.0) | < 0.001 |
| Any history of symptoms | |||||
| Buttock pain, n (%) | 54 (93.1) | 94 (97.9) | 148 (96.1) | 0 | 0.222 |
| Spinal pain, n (%) | 57 (98.3) | 89 (92.7) | 146 (94.8) | 0 | 0.195 |
| Arthralgia, n (%) | 47 (81.0) | 74 (78.9) | 122 (79.7) | 1 (0.6) | 0.796 |
| Peripheral arthritis, n (%) | 8 (13.8) | 14 (14.7) | 22 (14.4) | 1 (0.6) | 0.941 |
| Enthesopathy | |||||
| All sites, n (%) | 53 (91.4) | 78 (82.1) | 131 (85.6) | 1 (0.6) | 0.086 |
| Heel, n (%) | 37 (63.8) | 52 (54.7) | 89 (58.2) | 1 (0.6) | 0.176 |
| Anterior chest pain, n (%) | 43 (74.1) | 52 (54.7) | 95 (62.1) | 1 (0.6) | 0.012 |
| Dactylitis, n (%) | 4 (6.9) | 10 (10.5) | 14 (9.2) | 1 (0.6) | 0.544 |
| Anterior uveitis, n (%) | 2 (3.4) | 9 (9.5) | 11 (7.2) | 1 (0.6) | 0.213 |
| IBD, n (%) | 4 (6.9) | 15 (15.8) | 19 (12.4) | 1 (0.6) | 0.124 |
| Psoriasis, n (%) | 22 (37.9) | 16 (16.8) | 38 (24.8) | 1 (0.6) | 0.004 |
| Any history of biological inflammatory syndrome, n (%) | 23 (41.1) | 52 (54.7) | 75 (49.7) | 3 (1.9) | 0.053 |
| CRP at diagnosis, n > 5 (%) | 23 (40.4) | 39 (43.8) | 62 (42.8) | 9 (5.8) | 0.666 |
| CRP at diagnosis, mean ± SD (mg/l, among CRP > 5) | 30.6 ± 43.6 | 39.7 ± 52.2 | 36.4 ± 49.0 | 9 (5.8%) | 0.145 |
CRP C-reactive protein, IBD Chronic inflammatory bowel disease, n number, SD Standard deviation, SpA Spondyloarthritis
A lower rate of imaging joint damage was observed in the peripartum group compared to controls (Table 2), with structural damage on X-ray (at any sites, X-ray were carried out at the clinician’s discretion) found in 7.8% versus 20.3%, (p-value: 0.047), radiographic damage (sacroiliac joints) in 4% versus 18.6% (p-value: 0.012) and MRI sacroiliitis in 30.6% versus 47.4% (p-value: 0.046).
Table 2.
Radiographic and MRI involvement of women with SpA in the peripartum period and controls
| Peripartum SpA (n = 58) |
Controls (n = 96) |
p-value | |
|---|---|---|---|
| Radiographic damage at any site, n (%) | 4 (7.8) | 15 (20.3) | 0.047 |
| Radiographic damage of sacroiliac joint, n (%) | 2 (4.0) | 13 (18.6) | 0.012 |
| Sacro-iliitis on MRI at diagnosis, n (%) | 15 (30.6) | 36 (47.4) | 0.046 |
n number, SpA Spondyloarthritis
Classification criteria for women with SpA in the peripartum period and controls
The diagnosis of SpA had to be confirmed at the last follow-up visit. The mean duration of follow-up for all groups was 85.0 ± 64.7 months. Among the classification criteria (Table 3), 100% of our population met the ESSG criteria. Consistent with the findings on HLA-B27 and MRI sacroiliitis, only 38% of women in the peripartum group met the axial ASAS criteria, versus 65.6% of controls (p-value: 0.003). Similarly, the New York criteria were met by 3.8% versus 17.9% (p-value: 0.02) and 83.3% versus 99.0% for the Amor criteria (p-value < 0.001). Peripheral ASAS criteria were met in equal proportions in the peripartum group and controls,73.5% versus 71.6%. We found that 98.1% of women in the peripartum group met at least two sets of criteria among ESSG, ASAS, Amor and New York versus 99% of controls. Only one woman in each group met only one set of criteria.
Table 3.
Classification criteria of women with SpA in the peripartum period and controls
| SpA in the Peripartum period (n = 58) | Controls (n = 96) |
p-value | |
|---|---|---|---|
| Follow-up, mean ± SD (months) | 77.3 (59.5) | 89.6 (67.6) | |
| Axial ASAS criteria, n (%) | 19 (38.0) | 61 (65.6) | 0.003 |
| Peripheral ASAS criteria, n (%) | 36 (73.5) | 68 (71.6) | 0.847 |
| New York criteria, n (%) | 2 (3.8) | 12 (17.9) | 0.021 |
| Amor criteria, n (%) | 45 (83.3) | 95 (99.0) | < 0.001 |
| ESSG criteria, n (%) | 56 (100.0) | 96 (100.0) | 1.000 |
| At least 2 criteria, n (%) | 51 (98.1) | 95 (99.0) |
ASAS Assessment of SpondyloArthritis international Society, ESSG European spondylarthropathy study group, n number, SpA Spondyloarthritis
Treatment response of women with SpA in the peripartum period and controls
We identified 68 TNFα inhibitors in 32 women in the peripartum group and 123 TNFα inhibitors in 70 controls (Table 4). CRP level before TNFα inhibitor introduction, a known predictor of therapeutic response, was increased in 22.2% of women with peripartum SpA versus 48.1% of controls (p-value: 0.003). The delay between the onset of symptoms and TNFα inhibitor introduction was shorter in the peripartum group compared to the control group. Kaplan-Meier assessment of the retention rate of TNFα inhibitors at 12 months (Fig. 1A) shows a significant difference between groups: 45.6% (95% CI: 33.5–56.9) of women with peripartum SpA retained their treatment, compared with 71.2% (95% CI: 62.1–78.5) of controls (p-value < 0.001). This difference in treatment retention rate was already significant at 6 months, 67.6% (CI 95%: 55.1–77.4) in the peripartum group and 86.4% (CI 95%: 78.8–91.5) in the control group (p-value < 0.001). At 24 months, the retention rate of TNFα inhibitors was 26.5% (CI 95%: 16.7–37.3) in the peripartum group and 59.9% in controls (CI 95%: 50.5–68.2). The odds ratio for retaining TNFα inhibitors at 12 months according to the onset of symptoms in the peripartum period or not was 2.56 (CI 95%: 1.34–4.95).
Table 4.
Therapeutic evaluation of women with SpA in the peripartum period and controls
| Women with SpA in the peripartum period (n = 58) | Controls (n = 96) | p-value | |
|---|---|---|---|
| CRP before biotherapy, n > 5 (%) | 10 (22.2) | 38 (48.1) | 0.003 |
|
CRP before biotherapy, mean ± SD (mg/l among n > 5) |
16.1 ± 43.6 | 27.7 ± 34.6 | 0.333 |
| NSAID intake, n (%) | 42 (72.4) | 69 (75.0) | 0.713 |
| TNFα inhibitor therapy (n patients) | 68 (32) | 123 (70) | |
| BASDAI before treatment, mean ± SD | 6.74 ± 1.41 | 5.55 ± 1.53 | |
| BASFI before treatment, mean ± SD | 5.30 ± 2.00 | 4.66 ± 2.14 | |
| Time to initiation of bDMARD in months, mean (SD) | 63.3 ± 47.4 | 84.1 ± 63.3 | 0.033 |
| 6-months follow-up after biotherapy initiation | |||
| BASDAI50 responder, n (%) | 0 (0.0) | 11 (22.4) | |
| Low BASDAI activity, n (%) | 7 (13.7) | 28 (47.4) | |
| BASDAI improvement, mean ± SD | − 0.64 ± 2.04 | -1.13 ± 1.90 | |
| BASFI improvement, mean ± SD | − 0.40 ± 2.07 | -1.21 ± 2.25 | |
| Treatment retention rate at 12 months, n (%) | 31 (45.6) | 84 (68.3) | 0.002 |
No statistical tests were carried out on the BASDAI and BASFI data due to the high number of missing data.
n number, SD Standard deviation, SpA Spondyloarthritis
Fig. 1.
TNFα inhibitor retention rate for women with SpA in the peripartum period and controls. Comparison between women with SpA in the peripartum period and controls (A). Stratified representation on inflammatory sign (sacroiliitis on MRI and/or increased CRP before TNFα inhibitor introduction) and symptom onset period (B). Multivariate analysis of TNFα inhibitor retention rate based on sacroiliitis on MRI, increased CRP before TNFα inhibitor introduction and symptom onset period (C). CI: confidence interval; HR: hazard ratio; Spa: spondyloarthritis; *: p-value < 0.05
Before TNFα inhibitor introduction, BASDAI and BASFI scores were higher: 6.74 (standard deviation: ±1.41) in the peripartum group compared to controls, 6.74 ± 1.41 versus 5.55 ± 1.53, respectively. At 6-month follow-up, a worse treatment response was observed in women in the peripartum group with no evaluable patient reaching BASDAI50 (50% decrease in BASDAI) (Table 4) and only 13.7% reaching a low level of disease activity (defined as a BASDAI score of less than 4), compared with 47.4% of controls for whom data were available. The mean decreases in BASDAI and BASFI scores were respectively − 0.64 ± 2.04 and − 0.40 ± 2.07 in the peripartum group, and − 1.13 ± 1.90 and − 1.21 ± 2.25 in the control group. A stratified representation of retention rate on the presence of at least one objective inflammatory parameter (CRP greater than 5 mg/L prior to TNFα inhibitor introduction and/or sacroiliitis on MRI) shows an overlay of the curves for the inflammatory and non-inflammatory groups (Fig. 1B).
Finally, we performed a multivariate analysis of TNFα inhibitor discontinuation over 24 months using a Cox model. The variables analysed were a CRP level greater than 5 mg/L before TNFα inhibitor introduction or sacroiliitis on MRI, and peripartum SpA status (Fig. 1C). The hazard ratio (HR) was 0.78 (95% CI: 0.50–1.21) for MRI (p-value: 0.356) and 0.88 (95% CI: 0.58–1.32) for CRP (p-value: 0.544). Although not significant, there was a trend in our cohort for these two variables to favour retention rate. The HR for peripartum SpA status was 2.43 (95% CI: 1.62–3.65, p-value < 0.005).
Discussion
To our knowledge, this study is the first to describe the clinical, biological, and imaging features and the therapeutic response of women with SpA occurring in the peripartum period. We were able to identify in this population a weaker association with HLA-B27, more enthesitis and notably more anterior chest wall pain, less radiographic joint involvement and sacroiliitis on MRI compared with other women with SpA. These features meet the specificities of female SpA already described [12–15], which in our study seemed to predominate in the peripartum SpA group. The prevalence of HLA-B27 (72.2% in women from the DESIR cohort [15] - “devenir des spondylarthropathies indifférenciées récentes”, a French national SpA cohort) was lower in women with peripartum SpA than in controls (37.3% and 52.3% respectively). Unlike our cohort, the DESIR cohort only analyzed SpA with axial involvement. But since more than 98% of our patients had inflammatory back pain, comparison was reasonable. In contrast, anterior chest wall pain was involved in 44.6% of patients in the DESIR cohort [41] and in 54.7% of our controls. Its high prevalence in the peripartum SpA group (74.1%) might be explained by the histological structure of the manubrio-sternal joint. Indeed, this joint is the main site of anterior chest wall pain on MRI [42]. Together with the first rib, it forms a symphysis [43], composed of fibrocartilage rich in type I and III collagen [44] which are the main targets of metalloproteases responsible for cervical remodelling [24]. Serum levels of these metalloproteases in SpA are correlated with disease activity [45].
The association with psoriasis (17.1% in the DESIR cohort [41]), 16.8% in our controls, was markedly increased in the peripartum group (37.9%). Psoriasis is a highly IL-17-dependent skin disease, as demonstrated by the high efficacy of anti-IL17 in this condition [46]. In contrast, anti-IL17s have no proven efficacy in IBD or uveitis [47]. As mentioned above, IL17 may be one of the key factors driving the onset of disease in the peripartum period. We were unable to study the efficacy of anti-IL17 in our cohort and this will need further investigation.
The low prevalence of objective signs of inflammation, such as a sustained increase in CRP, or the absence of damage visible on imaging, may challenge the accuracy of the initial diagnosis. As pregnancy and parturition have a major physical and hormonal impact of pregnancy, it can trigger pain of various etiologies. Pregnancy-related pelvic pain, the origins of which are a complex combination of hormonal, biomechanical, metabolic and genetic actions, are grouped under the terminology “pregnancy-related pelvic girdle pain” [48] and highlights the significant changes in the patients’ musculoskeletal structures. The modification of the MRI signal of sacroiliacs joints from postpartum women, that can last up to two years, provides a good demonstration of these changes [49]. Traumatic pain, such as damage to the coccyx or ruptured symphysis [48], can also occur in the postpartum period and last for several months. However, careful clinical and iconographic analysis can help establish the diagnosis. Around 5% of such pain persists after childbirth [50]. These mechanical impairments may also contribute to the onset of SpA [5]. Fibromyalgia is frequently associated with SpA, especially in women, and may mimic its symptoms. Pregnancy and the postpartum have been described as exacerbating fibromyalgia [51], but we did not find any studies exploring its incidence in this period. The longer diagnosis delay in the peripartum SpA group could reflect the challenges in distinguishing these different diagnoses.
In our cohort, the biological inflammatory syndrome at diagnosis, the mean follow-up of 77.3 months and the positivity of at least two classification criteria in 98% of women with peripartum SpA (including ESSG, ASAS, New York and Amor criteria) are in favour of SpA. The association between the retention rate and the discontinuation of TNFα inhibitors and peripartum SpA status in the multivariate model (integrating sacroiliac MRI and CRP) is also a strong argument in favour of the importance of the latter. The low fulfillment of the ASAS criteria in the peripartum group could suggest that this set of classification is not appropriate for those SpA. As associations between peripheral enthesitis and anterior chest pain, the absence of HLA-B27, increased psoriasis and high BASDAI and BASFI scores were previously described, the phenotype of peripartum SpA seems to fit the same pattern [52]. The lack of difference in biological inflammatory syndrome at diagnosis, followed by a decrease in inflammation before TNFα inhibitor introduction in the peripartum group compared with controls may suggest an initial activation of the immune system, followed by a chronic sequellar phase. The activation of the immune system predominant in the initial phase of the disease could be another explanation for the lower retention rate of TNFα inhibitors.
The occurrence of SpA in the peripartum period was associated with a lower retention rate of TNFα inhibitors. At 12 months, the retention rate of women with peripartum SpA was 45.6% (Odds ratio: 2.56 [CI 95%: 1.34–4.95]) suggesting a low therapeutic response. Multivariate analysis using Cox models confirmed this negative association, finding a hazard ratio of 2.43 for the probability of TNFα inhibitor discontinuation in the first two years of treatment. Breast-feeding during the peripartum period may lead women to stop taking anti-TNFa. However, given the long time to diagnosis and introduction of biotherapy in our cohort, this was of only marginal importance. Sacroiliitis on MRI and elevated CRP, widely validated features of TNFα inhibitor success, could have their impact lowered in our female cohort where they are less common. On the contrary, as our cohort focused on peripartum SpA, the importance of this feature could be lowered in real practice depending on peripartum SpA incidence. In psoriatic arthritis, female sex was associated with a lower retention rate of TNFα inhibitors [17]. The retention rate after 2 years was 49.7% in men and 37.1% in women in the Danish Health Care Registers. In our study, the retention rate was 59.9% in controls and only 26.5% in women with SpA in the peripartum period after 2 years, which might suggest that removing the peripartum period could allow them to recover a maintenance level comparable with that of men. As treatments of less than one month were not analyzed, allergic reaction to TNFα inhibitor were not considered, and no adverse event occurred in our cohort: TNFα inhibitor discontinuation was due to inefficacy. This low retention rate could be due to a different pathophysiology in peripartum SpA, such as higher mechanical stress in enthesitis or IL-17 implication.
Our study also has limitations. As with any retrospective study, there can be much selection bias. The frequency of peripartum SpA cannot be assessed here. Moreover, because the onset of symptoms was often reported as a time period rather than a precise date, we were unable to analyze the interval between symptom onset and parturition. Our method of query using keywords probably inflated number of peripartum SpA compared to control. In a disease with potential intricated fibromyalgia symptoms, therapeutics response can be tricky to evaluate. Nevertheless, we made sure to exclude obvious diagnosis of fibromyalgia associated at diagnosis. The missing data (especially the WPI and SSS score) led us to use the ACR 2016 criteria for fibromyalgia, whose negativity was mainly determined by the following criteria in our cohort: “Generalized pain, defined as pain in at least 4 of 5 regions, must be present” and “Symptoms have been generally present for at least 3 months”. As a tertiary center specialized in auto immune disease, we decided to base SpA diagnosis on local clinician assumption. This decision made it possible to explore a wider range of clinical features, but could also increase the number of misdiagnosed patients. The amount of missing data also made it impossible to conclude on some results, such as treatment response assessed by BASDAI and BASFI scores. This response was nevertheless very low in women with peripartum SpA, as was the treatment retention rate.
Our work suggests that there is a distinct SpA phenotype in the peripartum period, with specific clinical features (more previous thoracic syndrome, more psoriasis), biological features (less biological inflammatory syndrome remote from diagnosis, a weaker association with HLA-B27), iconographic features (less radiographic and MRI damage) and therapeutic features (a lower retention rate of TNFα inhibitors). Peripartum onset could also be used as a negative predictor of treatment response. A better characterisation of SpA occurring in the peripartum period and of therapeutic response could lead to the development of specific treatments and contribute to the prevention of therapeutic wandering.
Acknowledgements
The authors are grateful to Nikki Sabourin-Gibbs, Rouen University Hospital, for her help in editing the manuscript. This study was presented as oral communication in the 36th congress of the “Société Française de Rhumatologie”.
Abbreviations
- BASFI
Bath Ankylosing Spondylitis
- BASDAI
Bath Ankylosing Spondylitis
- CI
confidence interval
- CRP
C reactive protein
- HR
hazard ratio
- IBD
inflammatory bowel disease
- N
number
- NSAID
non-steroidal anti-inflammatory drug
- SD
standard deviation
- SpA
Spondyloarthritis
Authors’ contributions
BdM, TL and OV designed the project. TL supervised the project. BdM, BG, PB, CPD, GA, NS, SP, SJD and JG collected and analyzed the data. BdM and TL wrote the manuscript with the help of all authors. All authors contributed to the article and approved the submitted version.
Funding
No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this article.
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
Declarations
Ethics approval and consent to participate
Our study was conducted in accordance with the Declaration of Helsinki and was approved by the local ethics committee (CSE – EDSaN) for research on existing data at Rouen University Hospital (2021/0190/OB). Moreover, the EDSaN data warehouse has been approved by the CNIL French agency (DT-2020-007, 19th october 2020). As we conducted a retrospective observational study using EDSaN (Rouen University Hospital’s clinical data warehouse), consent to participate was not mandatory (consent need was waived by the local ethics committee - CNIL French agency reference methodology MR004).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Watad A, Bridgewood C, Russell T, Marzo-Ortega H, Cuthbert R, McGonagle D. The Early Phases of Ankylosing Spondylitis: Emerging Insights From Clinical and Basic Science. Front Immunol. 2018;9:2668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Stolwijk C, Van Onna M, Boonen A, Van Tubergen A. Global Prevalence of Spondyloarthritis: A Systematic Review and Meta-Regression Analysis: Analysis of SpA Prevalence. Arthritis Care Res. 2016;68(9):1320–31. [DOI] [PubMed] [Google Scholar]
- 3.Navarro-Compán V, Sepriano A, El-Zorkany B, van der Heijde D. Axial spondyloarthritis. Ann Rheum Dis. 2021;80(12):1511–21. [DOI] [PubMed] [Google Scholar]
- 4.McGonagle D, Gibbon W, Emery P. Classification of inflammatory arthritis by enthesitis. Lancet. 1998;352(9134):1137–40. [DOI] [PubMed] [Google Scholar]
- 5.Jacques P, Lambrecht S, Verheugen E, Pauwels E, Kollias G, Armaka M, et al. Proof of concept: enthesitis and new bone formation in spondyloarthritis are driven by mechanical strain and stromal cells. Ann Rheum Dis. 2014;73(2):437–45. [DOI] [PubMed] [Google Scholar]
- 6.Noordenbos T, Yeremenko N, Gofita I, van de Sande M, Tak PP, Caňete JD, et al. Interleukin-17–positive mast cells contribute to synovial inflammation in spondylarthritis. Arthritis Rheum. 2012;64(1):99–109. [DOI] [PubMed] [Google Scholar]
- 7.Cañete JD, Celis R, Noordenbos T, Moll C, Gómez-Puerta JA, Pizcueta P, et al. Distinct synovial immunopathology in Behçet disease and psoriatic arthritis. Arthritis Res Ther. 2009;11(1):R17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Reinhardt A, Yevsa T, Worbs T, Lienenklaus S, Sandrock I, Oberdörfer L, et al. Interleukin-23–Dependent γ/δ T Cells Produce Interleukin-17 and Accumulate in the Enthesis, Aortic Valve, and Ciliary Body in Mice. Arthritis Rheumatol. 2016;68(10):2476–86. [DOI] [PubMed] [Google Scholar]
- 9.Martínez-Ramos S, Rafael-Vidal C, Pego-Reigosa JM, García S. Monocytes and Macrophages in Spondyloarthritis: Functional Roles and Effects of Current Therapies. Cells. 2022;11(3):515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Prati C, Lequerre T, Le Goff B, Cortet B, Toumi H, Tournadre A, et al. Novel insights into the anatomy and histopathology of the sacroiliac joint and correlations with imaging signs of sacroiliitis in case of axial spondyloarthritis. Front Physiol. 2023;14:1182902. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Abji F, Rasti M, Gómez-Aristizábal A, Muytjens C, Saifeddine M, Mihara K, et al. Proteinase-Mediated Macrophage Signaling in Psoriatic Arthritis. Front Immunol. 2021;11:629726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Tournadre A, Pereira B, Lhoste A, Dubost JJ, Ristori JM, Claudepierre P, et al. Differences Between Women and Men With Recent-Onset Axial Spondyloarthritis: Results From a Prospective Multicenter French Cohort: Sex Differences in Early Axial SpA. Arthritis Care Res. 2013;65(9):1482–9. [DOI] [PubMed] [Google Scholar]
- 13.Wright GC, Kaine J, Deodhar A. Understanding differences between men and women with axial spondyloarthritis. Semin Arthritis Rheum. 2020;50(4):687–94. [DOI] [PubMed] [Google Scholar]
- 14.Ortolan A, van Lunteren M, Ramiro S, Ramonda R, Landewé RBM, Dagfinrud H, et al. Are gender-specific approaches needed in diagnosing early axial spondyloarthritis? Data from the SPondyloArthritis Caught Early cohort. Arthritis Res Ther. 2018;20(1):218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Aouad K, Tournadre A, Lucasson F, Wendling D, Molto A, Fautrel B et al. Influence of Sex on Early Axial Spondyloarthritis: A Six-Year Longitudinal Analysis From a Large National Cohort. Arthritis Care Res. 2023;75(10):2107–16. [DOI] [PubMed]
- 16.Hoballah A, Lukas C, Leplat C, Taourel P, Pialat JB, Sans N, et al. MRI of sacroiliac joints for the diagnosis of axial SpA: prevalence of inflammatory and structural lesions in nulliparous, early postpartum and late postpartum women. Ann Rheum Dis. 2020;79(8):1063–9. [DOI] [PubMed] [Google Scholar]
- 17.Højgaard P, Ballegaard C, Cordtz R, Zobbe K, Clausen M, Glintborg B et al. Gender differences in biologic treatment outcomes—a study of 1750 patients with psoriatic arthritis using Danish Health Care Registers. Rheumatology. 2018;57(9):1651–60. [DOI] [PubMed] [Google Scholar]
- 18.McHugh NJ, Laurent MR. The effect of pregnancy on the onset of psoriasitic arthritis. Rheumatology. 1989;28(1):50–2. [DOI] [PubMed] [Google Scholar]
- 19.Ostensen M, Ostensen H. Ankylosing spondylitis–the female aspect. J Rheumatol. 1998;25(1):120–4. [PubMed] [Google Scholar]
- 20.Wallenius M, Skomsvoll JF, Irgens LM, Salvesen KÅ, Koldingsnes W, Mikkelsen K, et al. Postpartum onset of rheumatoid arthritis and other chronic arthritides: results from a patient register linked to a medical birth registry. Ann Rheum Dis. 2010;69(2):332–6. [DOI] [PubMed] [Google Scholar]
- 21.Hamroun S, Hamroun A, Bigna JJ, Allado E, Förger F, Molto A. Fertility and pregnancy outcomes in women with spondyloarthritis: a systematic review and meta-analysis. Rheumatology. 2022;61(4):1314–27. [DOI] [PubMed] [Google Scholar]
- 22.Meissner Y, Strangfeld A, Molto A, Forger F, Wallenius M, Costedoat-Chalumeau N et al. Pregnancy and neonatal outcomes in women with axial spondyloarthritis: pooled data analysis from the European Network of Pregnancy Registries in Rheumatology (EuNeP). Ann Rheum Dis. 2022;81(11):1524–33. [DOI] [PubMed]
- 23.Ursin K, Lydersen S, Skomsvoll JF, Wallenius M. Disease activity during and after pregnancy in women with axial spondyloarthritis: a prospective multicentre study. Rheumatology. 2018;57(6):1064–71. [DOI] [PubMed] [Google Scholar]
- 24.Yellon SM. Immunobiology of Cervix Ripening. Front Immunol. 2020;10:3156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Calguneri M, Bird HA, Wright V. Changes in joint laxity occurring during pregnancy. Ann Rheum Dis. 1982;41(2):126–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Fiani B, Sekhon M, Doan T, Bowers B, Covarrubias C, Barthelmass M, et al. Sacroiliac Joint and Pelvic Dysfunction Due to Symphysiolysis in Postpartum Women. Cureus. 2021;13(10):e18619. [DOI] [PMC free article] [PubMed]
- 27.Varma S, Orgel JPRO, Schieber JD. Nanomechanics of Type I Collagen. Biophys J. 2016;111(1):50–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Gomez-Lopez N, StLouis D, Lehr MA, Sanchez-Rodriguez EN, Arenas-Hernandez M. Immune cells in term and preterm labor. Cell Mol Immunol. 2014;11(6):571–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ditzian-Kadanoff R, Garon J, Verp MS, Zilberstein M. γδ T cells in human decidua. Am J Obstet Gynecol. 1993;168(3):831–6. [DOI] [PubMed] [Google Scholar]
- 30.Pinget GV, Corpuz TM, Stolp J, Lousberg EL, Diener KR, Robertson SA, et al. The majority of murine γδ T cells at the maternal–fetal interface in pregnancy produce IL-17. Immunol Cell Biol. 2016;94(7):623–30. [DOI] [PubMed] [Google Scholar]
- 31.Pavlov O, Pavlova O, Ailamazyan E, Selkov S. Characterization of cytokine production by human term placenta macrophages in vitro. Am J Reprod Immunol N Y N 1989. 2008;60(6):556–67. [DOI] [PubMed] [Google Scholar]
- 32.Pressat-Laffouilhère T, Balayé P, Dahamna B, Lelong R, Billey K, Darmoni SJ, et al. Evaluation of Doc’EDS: a French semantic search tool to query health documents from a clinical data warehouse. BMC Med Inf Decis Mak. 2022;22(1):34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Häuser W, Katz RL, et al. 2016 Revisions to the 2010/2011 fibromyalgia diagnostic criteria. Semin Arthritis Rheum. 2016;46(3):319–29. [DOI] [PubMed] [Google Scholar]
- 34.Maksymowych WP, Lambert RG, Østergaard M, Pedersen SJ, Machado PM, Weber U, et al. MRI lesions in the sacroiliac joints of patients with spondyloarthritis: an update of definitions and validation by the ASAS MRI working group. Ann Rheum Dis. 2019;78(11):1550–8. [DOI] [PubMed] [Google Scholar]
- 35.Rudwaleit M, van der Heijde D, Landewé R, Akkoc N, Brandt J, Chou CT, et al. The Assessment of SpondyloArthritis international Society classification criteria for peripheral spondyloarthritis and for spondyloarthritis in general. Ann Rheum Dis. 2011;70(1):25–31. [DOI] [PubMed] [Google Scholar]
- 36.Linden SVD, Valkenburg HA, Cats A. Evaluation of Diagnostic Criteria for Ankylosing Spondylitis. Arthritis Rheum. 1984;27(4):361–8. [DOI] [PubMed] [Google Scholar]
- 37.Amor B, Dougados M, Mijiyawa M. Criteria of the classification of spondylarthropathies. Rev Rhum Mal Osteoartic. 1990;57(2):85–9. [PubMed] [Google Scholar]
- 38.Dougados M, Linden SVD, Juhlin R, Huitfeldt B, Amor B, Calin A, et al. The European Spondylarthropathy Study Group Preliminary Criteria for the Classification of Spondylarthropathy. Arthritis Rheum. 1991;34(10):1218–27. [DOI] [PubMed] [Google Scholar]
- 39.Klein JP, Logan B, Harhoff M, Andersen PK. Analyzing survival curves at a fixed point in time. Stat Med. 2007;26(24):4505–19. [DOI] [PubMed] [Google Scholar]
- 40.Lukas C, Dougados M, Combe B. Factors associated with a bad functional prognosis in early inflammatory back pain: results from the DESIR cohort. RMD Open. 2016;2(1):e000204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Wendling D, Prati C, Demattei C, Loeuille D, Richette P, Dougados M. Anterior chest wall pain in recent inflammatory back pain suggestive of spondyloarthritis. data from the DESIR cohort. J Rheumatol. 2013;40(7):1148–52. [DOI] [PubMed] [Google Scholar]
- 42.Weber U, Lambert RG, Rufibach K, Maksymowych WP, Hodler J, Zejden A, et al. Anterior chest wall inflammation by whole-body magnetic resonance imaging in patients with spondyloarthritis: lack of association between clinical and imaging findings in a cross-sectional study. Arthritis Res Ther. 2012;14(1):R3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sarcinelli JM, Eustáquio-Silva R, da Silva Baptista J. Manubriosternal joint: synchondrosis or symphysis? Analysis of morphology and aging in humans. Surg Radiol Anat. 2019;41(3):275–9. [DOI] [PubMed] [Google Scholar]
- 44.Pirozzi KM. Histophysiology of Fibrocartilage. Clin Podiatr Med Surg. 2022;39(3):363–70. [DOI] [PubMed] [Google Scholar]
- 45.Gudmann NS, Siebuhr AS, Christensen AF, Ejstrup L, Sørensen GL, Loft AG, et al. Type I and III collagen turnover is increased in axial spondyloarthritis and psoriatic arthritis. Associations with disease activity and diagnostic capacity. Clin Exp Rheumatol. 2017;35(4):653–9. [PubMed] [Google Scholar]
- 46.Armstrong AW, Read C, Pathophysiology. Clinical Presentation, and Treatment of Psoriasis: A Review. JAMA. 2020;323(19):1945. [DOI] [PubMed] [Google Scholar]
- 47.Webers C, Ortolan A, Sepriano A, Falzon L, Baraliakos X, Landewé RBM, et al. Efficacy and safety of biological DMARDs: a systematic literature review informing the 2022 update of the ASAS-EULAR recommendations for the management of axial spondyloarthritis. Ann Rheum Dis. 2023;82(1):130–41. [DOI] [PubMed] [Google Scholar]
- 48.Kanakaris NK, Roberts CS, Giannoudis PV. Pregnancy-related pelvic girdle pain: an update. BMC Med. 2011;9:15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Renson T, Depicker A, De Craemer AS, Deroo L, Varkas G, de Hooge M, et al. High prevalence of spondyloarthritis-like MRI lesions in postpartum women: a prospective analysis in relation to maternal, child and birth characteristics. Ann Rheum Dis. 2020;79(7):929–34. [DOI] [PubMed] [Google Scholar]
- 50.Norén L, Östgaard S, Johansson G, Östgaard HC. Lumbar back and posterior pelvic pain during pregnancy: a 3-year follow-up. Eur Spine J. 2002;11(3):267–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Østensen M, Rugelsjoen A, Wigers SH. The Effect of Reproductive Events and Alterations of Sex Hormone Levels on the Symptoms of Fibromyalgia. Scand J Rheumatol. 1997;26(5):355–60. [DOI] [PubMed] [Google Scholar]
- 52.Nadon V, Moltó A, Etcheto A, Bessette L, Michou L, D’Agostino MA, et al. Clinical peripheral enthesitis in the DESIR prospective longitudinal axial spondyloarthritis cohort. Clin Exp Rheumatol. 2019;37(4):561–5. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.